Childlessness and Intergenerational Transfers in Later Life
Childlessness in later life has been attracting increased attention from researchers and policy makers. Yet a number of misconceptions about childlessness among the elderly remain, such as the claim that elderly childless people are mainly on the receiving end of intergenerational exchanges, or that they are a homogeneous group. Contrary to these assumptions, we find that elderly childless people give as well as receive, and that parental status is a continuum, ranging from full childlessness across several intermediary conditions to full current natural parenthood. In a study of the elderly population across 11 European countries, we show that non-parents make significant contributions to their social networks of family and friends through financial and time transfers, and that the latter in particular differ little from those of natural parents. The same applies to their participation in charitable and voluntary work. Different parental statuses are significantly associated with the various dimensions of giving and receiving. Social parents (i.e., people who have no natural children, but who have adopted, foster, or stepchildren) are shown to be much more similar to natural parents than to non-parents. Family recomposition thus does not seem to inhibit intergenerational exchanges, as long as social parents have sufficient contact with their non-natural social children. On the other hand, parents who have lost contact with their children – natural or otherwise – are likely to require more formal care in later life.
KeywordsChildlessness Aging Intergenerational support Intergenerational transfers Step families
After reaching a low point among the 1935–1945 birth cohort, childlessness has increased significantly in recent decades in most European societies (Rowland 2007; OECD 2010; Tanturri et al. 2015). In previous research on childlessness, a recurring theme has been the consequences for an individual’s risk of social isolation and insufficient informal support, particularly in later life (Kohli and Albertini 2009). From the perspective of public policy, childless elderly people are usually seen as a problem group. It has been shown that parent-child relations are central to the social embeddedness of elderly people. Thus, it is generally assumed that compared to adults who have children, childless adults are at higher risk of lacking the social and emotional support they will need when they become frail and dependent. Citing the negative effects of the absence of children on social inclusion, policy makers have expressed concerns that increasing rates of childlessness among the elderly population will lead to increasing demands for public social care and health services.
There are, however, two reasons why this assumption may be flawed. First, childless elderly people are not only on the receiving end of support; they also give to their families and to society at large by establishing strong linkages with next-of-kin relatives, investing in non-family networks, and participating in voluntary and charitable activities. Taking these transfers and activities into account, we have found that the differences in the support exchange behaviours between parents and childless adults are small (Albertini and Kohli 2009). Second, childless elderly people are not a homogenous group. Childlessness should be seen as a life course process across a series of decision and bifurcation points (Kreyenfeld and Konietzka 2007). The social consequences of being childless in later life depend on the specific paths into childlessness (Dykstra and Hagestad 2007; Keizer et al. 2010; Mynarska et al. 2015), and they may also depend on the specific family and kinship constellations of each childless individual.
The aim of the present chapter is to address these two points. We report the results of a new study that deals with the social consequences of childlessness in later life by looking at the support given and received, and that examines parenthood and childlessness not as two exclusive alternatives, but as a continuum across a range of intermediate statuses. Thus, we analyse not only the financial and social support childless elderly people receive, but also the support they provide to their kin and friends, and to the society in which they live; and we map the patterns of support onto the different types of parental and childlessness status.
17.2 Social Consequences of Childlessness: Patterns of Support
The social consequences of childlessness in old age are multiple and complex. They vary with the specific institutional setting, and, at the individual level, with the specific motivation for and the pathway to childlessness. How someone ends up with no children may be more important than not having a child per se. Choosing not to have children, being unable to find a partner, not being fecund, surviving the death of one’s children, and being socially childless because of early divorce represent different paths to childlessness, and each of these paths has different connotations. Marital history and gender also mediate the consequences of childlessness for individuals, as do the usual cleavages of education, income, and health.
Raising children requires the investment of substantial financial and time resources by parents, and there is a general recognition that the costs associated with parenthood outweigh the benefits, at least while children are young (for a literature review on the costs of children see Folbre 2008). At the same time, research on well-being in old age has shown that adult children have a positive impact overall on parents’ well-being (for a review of studies on parenthood and well-being over the last decade, see Umberson et al. 2010) and even on mortality: People tend to live longer if they have a surviving adult child. This effect of children on life expectancy is mediated by people’s perceptions of the emotional and social support that is available to them in case of need. The effect also extends to parents who have survived, abandoned, or lost contact with their children (Weltoft et al. 2004). One explanation for parents’ higher life expectancy may be the healthier behaviour that parenthood encourages (Dykstra and Hagestad 2007).
According to an influential theory of the modern transition to low fertility, one of the main reasons why people had children in the past was because the children were expected to provide social and economic support when the parents became old and frail and were no longer able to be self-sufficient (Caldwell 1976); whereas today older people no longer depend on the support of their descendants in old age because they can now rely on pensions, health care, and social services provided by the welfare state (Nuget 1985). Some authors have argued that such old-age security motives for having children – ensuring material support and care in old age – still apply today, not just in low-welfare developing societies, but to some extent also in affluent societies with extensive welfare states (Kreager and Schröder-Butterfill 2004; Boldrin et al. 2005). While this controversy has yet to be resolved, it has been documented that elderly people in affluent societies continue to be embedded in dense intergenerational family networks of support, especially between parents and their children (Albertini et al. 2007; Kohli et al. 2010). Apart from providing direct support, children can serve as important intermediaries between their parents and health and social care services, and can thus help their parents gain access to the public resources available to the aged population (Choi 1994).
Given that adult children continue to represent an important source of support for elderly parents, we may assume that childless older people have a higher risk than parents of lacking social and moral support when they become frail and dependent. The evidence to date only partially confirms this expectation. Generally, the childless do not appear to have larger support deficits than parents (Albertini and Mencarini 2014). Childless people tend to compensate for the absence of exchanges with adult children by having frequent contact with neighbours and friends, and by developing strong ties with other family members, including with their parents, their siblings, and their nephews and nieces (Albertini and Kohli 2009; Schnettler and Woehler 2015). Moreover, despite the stigma that may still be attached to voluntary childlessness and the distress that may accompany involuntary childlessness (Dykstra and Hagestad 2007), recent empirical evidence does not support the assumption that childless older people have lower levels of economic, psychological, or social well-being than their counterparts who have children (Hank and Wagner 2013).
However, the evidence also indicates that when intensive support is needed, these compensatory strategies work only partially. When they become frail and limited in their ability to carry out the activities of daily living, childless people receive less support and are more likely to enter residential care, and do so at lower levels of dependency compared with people who have children (Wenger 2009). If the share of the childless population increases, we may expect that the share of those who lack family support – and thus the demand for public health and social care services – will also grow. Given the constraints on welfare state spending, it is possible that this additional demand will not be met, and that childless older people will have to look to the private market for alternative solutions. Even in an advanced welfare state such as Sweden, public home help services have not been able to fully compensate for the lack of family support among the childless (Larsson and Silverstein 2004).
At the same time, however, the debate about the effects of increasing levels of childlessness on the future demand for social care has neglected the opposite flows of support: How the absence of children affects what older people give. Contrary to widespread perceptions, on balance elderly people make more transfers and provide more support than they receive (Kohli et al. 2010). We have shown that although childless elderly people are less likely than parents to provide financial transfers and social support to others, these transfers and supports are still substantial (Albertini and Kohli 2009). A study conducted in the United States found that compared with parents, childless older people are more likely to make financial transfers to other kin, friends, and neighbours; and that they transfer larger amounts (Hurd 2009). A considerable share of these transfers still go to descendants such as nephews and nieces, and can therefore be considered intergenerational giving. Moreover, because they have a greater need to construct social networks outside of their families, childless people may be expected to give more of their time and money to charitable and community activities, and thus contribute more to society at large. Hurd (2009) shows that childless older people in the U.S. indeed donated larger amounts of money to charities than parents. To the extent that these organizations focus on young people, this type of giving is again intergenerational.
17.3 Parenthood as a Continuum
As we noted above, a large body of previous research on childless people has treated non-parents and parents as two homogeneous groups, distinguishing only between those who had and those who did not have living children at the time of the interview. There is, however, increasing evidence that there are different pathways to childlessness, and that the consequences of childlessness vary depending on these pathways and their endpoints. The same is true for parents. There is no straightforward distinction between being or not being a parent: a person can become a parent as the result of having a natural child (with or without the help of assisted reproduction technologies), but also by adopting a child or becoming a stepparent of a partner’s child. Thus, people can have children through different routes and at different points in their life course. A person can also cease to be a parent. The most obvious case in which this occurs is when a parent has survived his/her children. But there are also parents who, due to life events such as a divorce or an intense family conflict, have lost track of their children and no longer have contact with them. Other parents have children who live very far away (see Schnettler and Woehler 2015). These situations may have different effects on support networks and exchanges. Our empirical analysis is a first step towards taking these different situations into account. We distinguish between those who have natural children and stay in contact with them; those who have had natural children but have survived them, have lost contact with them, or live far away from them; those who did not have natural children but have adopted, foster, or stepchildren; and, finally, those who never had any children, natural or otherwise. Thus, we conceptualise parenthood and childlessness not as two fully separate conditions, but as a continuum of parental statuses.
17.4 Analytic Approach, Data, and Variables
The data for this analysis is drawn from the first three regular waves of the Survey of Health, Ageing and Retirement in Europe (SHARE) collected in 2004, 2007, and 2011; and from the retrospective third wave (SHARELIFE) collected in 2009. We use data from the 11 European countries that participated in the first wave of SHARE: Austria, Belgium, Denmark, France, Germany, Greece, Italy, the Netherlands, Spain, Sweden, and Switzerland.
SHARE is a longitudinal, cross-national survey representative of the population aged 50 and older; the partners of the respondents (regardless of their age) are also included. SHARE contains detailed information on the financial transfers and social support (including formal and informal care) given and received during the 12 months prior to each interview. Using the combined information of SHARE and SHARELIFE, we were able to distinguish between different types of parenthood and childlessness. We created six subgroups of respondents: (a) those who never had natural children and had no adopted, foster, or stepchildren at the time of the interview (fully childless); (b) those who had natural children, but no living children at the time of the interview (survived all children); (c) those who never had natural children, but who at the time of the interview had adopted, foster, or stepchildren who were living less than 500 km away with whom they had contact at least once a month (social parents); (d) those who had natural children, and who at the time of the interview still had at least one child who was living less than 500 km away with whom they had contact at least once a month (natural parents); (e) those who had at least one living child at the time of the interview (natural, step, adopted, or foster), but who had lost contact with all of their children (i.e., less than one contact per month or no contact at all during the 12 months prior to the interview) (parents no contact); and (f) those who had at least one living child at the time of the interview (natural, step, adopted, or foster), but who were living more than 500 km away from their nearest child (parents geographical distance).
Sample characteristics, column per cent
Survived all children
Parents who have lost contact with children
Parents who live at >500 km away from children
Married or in registered partnership
None (ISCED 0)
Low (ISCED 1 & 2)
Intermediate (ISCED 3 & 4)
High (ISCED 5 & 6)
Has at least one limitation
Age (mean, SD)
Household equivalent income (ppp), (mean, SD)
Household per-capita wealth (ppp), (mean, SD)
First, we report some descriptive statistics on the support networks of the six types of parents/non-parents. The second step consists of multivariate analyses of support exchange. The previous literature has consistently shown that elderly parents and non-parents differ systematically in their characteristics, such as economic resources, health, and partnership status. These characteristics are also important factors that influence personal support networks. Therefore, in order to analyse the relationship between parental status and support exchange, we need to control for a number of possible compositional effects. We introduce the following control variables into our multivariate analyses: age, marital status (i.e., married or in a registered partnership, separated or divorced, widowed, never married), educational level (measured according to the ISCED-97 scale), health status (measured as the presence of at least one limitation on the Global Activity Limitation Index [GALI], or on the Activities of Daily Living [ADL] or Instrumental Activities of Daily Living [IADL] indicators, or on the indicator of mobility and fine motor limitations), the natural logarithm of household equivalent income, household net per capita wealth, and the country of residence.
The multivariate analyses are carried out by using population-averaged logit and linear regression models for binomial and continuous variables, respectively, on the unbalanced sample of respondents taking part in at least one of the first three regular waves of SHARE. We consider several dependent variables: the likelihood of giving/receiving social support (i.e., help with paperwork, household chores, personal care) to/from non-coresiding individuals; the natural logarithm of the amount of social support given/received expressed as the estimated number of hours per year (this variable is only available for the first two waves of the survey); the likelihood of giving/receiving financial support to/from others; the likelihood of participating in the activities of charitable or voluntary organizations in the 4 weeks prior to the interview, and the likelihood of providing this support on a weekly or daily basis (these variables are only available for the first two waves of the survey); and the likelihood of receiving professional or paid home help, or of staying overnight in a nursing home in the last 12 months (this variable is only available for the first two waves of the survey). Because the previous literature has shown that the lack of children has different effects for men and women, we estimate separate models for these two groups. Due to space limitations we report below only the regression coefficients for the different parental statuses, while omitting those for the controlling variables.1
Characteristics of the respondents’ support network by parental status
Survived all children
Parents no contact
Parents geo distance
% Giving economic support
% Giving social support
Mean amount of social support given
% Participating in charitable or voluntary work
% Participating in charitable or voluntary work on a daily or weekly basis
% Receiving economic support
% Receiving social support
Mean amount of social support received
% Receiving professional or paid care support (home care or nursing home)
Moving the focus to the support received, Table 17.2 indicates that contrary to expectations, non-natural parents and parents who had lost contact with their children were more likely than natural parents to have been receiving social support. These groups, together with the group of parents who were living more than 500 km away from their children, were also more likely to have been receiving formal care support.
Clearly, all of these differences between the types of parenthood or childlessness could be the result of systematic compositional differences. For instance, natural and social parents might, on average, be younger and/or in better economic and health conditions than the other respondents, and these differences could explain why they were less likely to be receiving formal and informal social support. For this reason, the next step of our analysis is to investigate the relationship between childlessness and support networks in a multivariate framework.2
17.5.1 What Childless People Give
One of the most overlooked topics in the study of childless elderly people is the extent to which they contribute to others (relatives and non-relatives) and to society at large. Most of the previous research on elderly non-parents has focused on the challenges they face later in life. As we have shown (Albertini and Kohli 2009), however, the amount of support provided by non-parents to others is far from negligible. In the present section, we want to address this issue by examining the contributions of elderly people based on their parental status.
In sum, these results show that the likelihood of financial support to others is clearly associated with having or not having children, and that for fathers whether they had regular contact with their children is also a factor. Generally, the fully childless, those who had survived their children, and those who had lost contact with them are less likely to have been making financial transfers than parents. It seems that the two latter groups of fathers are located between the two extremes of the financial transfer behaviour of natural fathers and fully childless men. In contrast, social support is less clearly connected with the presence of children, except among fully childless women and mothers who had lost contact with their children. The results for participation in charitable or voluntary work are similar: while we find little evidence that the childless were playing a special role in these forms of social engagement beyond their immediate circle of family and friends, our findings do contradict the common assumption that childless people are ego-centred and isolated members of contemporary societies.
17.5.2 What Childless People Receive
As was mentioned above, most previous research on the social networks of the childless has focused on what they lack in terms of informal social support. Here we complement this approach by including in our analysis both the formal and the informal types of support the childless receive, and by investigating how the levels and the types of support they receive differ across the range of parental statuses.
 The latter finding confirms the results of a recent study of the elderly Dutch population by van der Pers et al. (2015) which showed that having children living close by was negatively associated with the likelihood of moving to a care institution.
Childlessness in later life is a topic that has been attracting increased levels of attention from researchers and policy makers. It is also still the subject of widely held misconceptions. Two of the most misleading ones are that childless elderly people are only or mainly at the receiving end of intergenerational exchanges, and that they are all of one kind. Contrary to these assumptions, we find that elderly childless people give as well as receive support, and that parental status is a continuum, ranging from full childlessness across several intermediary conditions to full current natural parenthood.
In a study of the elderly population across 11 European countries, we have shown that non-parents make significant contributions to their social networks of family and friends through financial and time transfers, and that their contributions of time in particular differ little from those of natural parents. The same applies to participation in charitable and voluntary work. Different parental statuses are significantly associated with the various dimensions of giving and receiving. The patterns across these dimensions and statuses need to be examined in detail, but two general results stand out. The first is that social parents (i.e., people who have no natural children but who have adopted, foster, or stepchildren) are more similar to natural parents than to non-parents. Family recomposition thus does not seem to inhibit intergenerational exchanges as long as social parents have sufficient contact with their social children. The second result is that parents who have lost contact with their children – natural or otherwise – are an overlooked group in terms of their heightened demand for formal care in later life. As this group may be increasing in size, it represents a special challenge for policy.
- Folbre, N. (2008). Valuing children. Cambridge, MA: Harvard University Press.Google Scholar
- Kohli, M., Albertini, M., & Künemund, H. (2010). Linkages among adult family generations: Evidence from comparative survey research. In P. Heady & M. Kohli (Eds.), Family, kinship and state in contemporary Europe (Perspectives on theory and policy, Vol. 3, pp. 195–220). Frankfurt am Main: Campus.Google Scholar
- Kreager, P., & Schröder-Butterfill, E. (Eds.). (2004). Ageing without children: European and Asian perspectives. Oxford: Berghahn.Google Scholar
- OECD. (2010). OECD family database Paris. OECD – Social Policy Division – Directorate of Employment, Labour and Social Affairs. SF2.5: Childlessness.Google Scholar
- van der Pers, M., Mulder, C. H., & Steverink, N. (2015). Geographic proximity of adult children and the well-being of older persons. Research on Aging, 37, 524–551.Google Scholar
- Schnettler, S., & Woehler, T. (2015). No children in later life, but more and better friends? Substitution mechanisms in the personal and support networks of parents and the childless in Germany. Ageing and Society, pre-publication view.Google Scholar
- Tanturri, M. L., Mills, M., Rotkirch, A., Sobotka, T., Takacs, J., Miettinen, A., Faludi, C., Kantsa, V., & Nasiri, D. (2015). State-of-the-art report. Childlessness in Europe (Families and societies working paper series, 32).Google Scholar
Open Access This chapter is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the work’s Creative Commons license, unless indicated otherwise in the credit line; if such material is not included in the work’s Creative Commons license and the respective action is not permitted by statutory regulation, users will need to obtain permission from the license holder to duplicate, adapt or reproduce the material.